Blog tags

 Subscribe to new blog posts.

• significance 11
• power 12
• simplicity 9
• R 33
• breakpoint regression 1
• multiple comparisons 5
• silly tests 4
• graphics 13
• effect sizes 11
• measurement error 5
• design features 7
• organisation 4
• logistic regression 4
• correlational studies 3
• mixed-effects models 3
• cluster-randomised experiments 2
• generalised additive models 2
• non-linearities 4
• missing data 1
• open science 2
• correction 1
• multiple regression 3
• reading 1
• tutorial 6
• machine learning 1
• random forests 1
• R tip 1
• bootstrapping 1
• replication 1
• predictive modelling 1

significance

• Confidence interval-based optional stopping
• The Centre for Open Science's Preregistration Challenge: Why it's relevant and some recommended background reading
• On correcting for multiple comparisons: Five scenarios
• Experiments with intact groups: spurious significance with improperly weighted t-tests
• Analysing experiments with intact groups: the problem and an easy solution
• Power simulations for comparing independent correlations
• Explaining key concepts using permutation tests
• Assessing differences of significance
• A purely graphical explanation of p-values
• Calibrating p-values in 'flexible' piecewise regression models
• Analysing pretest/posttest data

power

• Abandoning standardised effect sizes and opening up other roads to power
• On correcting for multiple comparisons: Five scenarios
• Causes and consequences of unequal sample sizes
• The problem with cutting up continuous variables and what to do when things aren't linear
• Analysing experiments with intact groups: the problem and an easy solution
• Covariate adjustment in logistic mixed models: Is it worth the effort?
• Covariate adjustment in logistic regression — and some counterintuitive findings
• Silly significance tests: Tests unrelated to the genuine research questions
• Power simulations for comparing independent correlations
• More on why I don't like standardised effect sizes
• The curious neglect of covariates in discussions of statistical power
• Analysing pretest/posttest data

simplicity

• Common-language effect sizes
• Surviving the ANOVA onslaught
• Silly significance tests: The main effects no one is interested in
• Silly significance tests: Tests unrelated to the genuine research questions
• A more selective approach to reporting statistics
• Overaccuracy and false precision
• Silly significance tests: Tautological tests
• Silly significance tests: Balance tests
• Analysing pretest/posttest data

R

• Confidence interval-based optional stopping
• Creating comparable sets of stimuli
• Interactions between continuous variables
• Tutorial: Adding confidence bands to effect displays
• Tutorial: Plotting regression models
• Confidence intervals for standardised mean differences
• Some illustrations of bootstrapping
• What data patterns can lie behind a correlation coefficient?
• Common-language effect sizes
• Tutorial: Drawing a dot plot
• R tip: Ordering factor levels more easily
• Classifying second-language learners as native- or non-nativelike: Don't neglect classification error rates
• Tutorial: Drawing a boxplot
• Tutorial: Drawing a line chart
• Tutorial: Drawing a scatterplot
• Why reported R² values are often too high
• Experiments with intact groups: spurious significance with improperly weighted t-tests
• Drawing a scatterplot with a non-linear trend line
• Causes and consequences of unequal sample sizes
• The problem with cutting up continuous variables and what to do when things aren't linear
• Analysing experiments with intact groups: the problem and an easy solution
• Covariate adjustment in logistic mixed models: Is it worth the effort?
• Covariate adjustment in logistic regression — and some counterintuitive findings
• Some tips on preparing your data for analysis
• Power simulations for comparing independent correlations
• Explaining key concepts using permutation tests
• Thinking about graphs
• Some alternatives to bar plots
• Assessing differences of significance
• Silly significance tests: Balance tests
• A purely graphical explanation of p-values
• Calibrating p-values in 'flexible' piecewise regression models
• Analysing pretest/posttest data

breakpoint regression

• Calibrating p-values in 'flexible' piecewise regression models

multiple comparisons

• The Centre for Open Science's Preregistration Challenge: Why it's relevant and some recommended background reading
• Why reported R² values are often too high
• On correcting for multiple comparisons: Five scenarios
• Silly significance tests: Tests unrelated to the genuine research questions
• Calibrating p-values in 'flexible' piecewise regression models

silly tests

• Silly significance tests: The main effects no one is interested in
• Silly significance tests: Tests unrelated to the genuine research questions
• Silly significance tests: Tautological tests
• Silly significance tests: Balance tests

graphics

• Interactions between continuous variables
• Tutorial: Adding confidence bands to effect displays
• Tutorial: Plotting regression models
• What data patterns can lie behind a correlation coefficient?
• Tutorial: Drawing a dot plot
• R tip: Ordering factor levels more easily
• Tutorial: Drawing a boxplot
• Tutorial: Drawing a line chart
• Tutorial: Drawing a scatterplot
• Drawing a scatterplot with a non-linear trend line
• A more selective approach to reporting statistics
• Thinking about graphs
• Some alternatives to bar plots

effect sizes

• Abandoning standardised effect sizes and opening up other roads to power
• Confidence intervals for standardised mean differences
• Which predictor is most important? Predictive utility vs. construct importance
• What data patterns can lie behind a correlation coefficient?
• Common-language effect sizes
• Why reported R² values are often too high
• Covariate adjustment in logistic mixed models: Is it worth the effort?
• Covariate adjustment in logistic regression — and some counterintuitive findings
• More on why I don't like standardised effect sizes
• A more selective approach to reporting statistics
• Why I don't like standardised effect sizes

measurement error

• Abandoning standardised effect sizes and opening up other roads to power
• Which predictor is most important? Predictive utility vs. construct importance
• Controlling for confounding variables in correlational research: Four caveats
• More on why I don't like standardised effect sizes
• Why I don't like standardised effect sizes

design features

• Confidence interval-based optional stopping
• Creating comparable sets of stimuli
• Abandoning standardised effect sizes and opening up other roads to power
• Experiments with intact groups: spurious significance with improperly weighted t-tests
• Causes and consequences of unequal sample sizes
• Analysing experiments with intact groups: the problem and an easy solution
• Explaining key concepts using permutation tests

organisation

• Creating comparable sets of stimuli
• Automatise repetitive tasks
• The Centre for Open Science's Preregistration Challenge: Why it's relevant and some recommended background reading
• Some tips on preparing your data for analysis

logistic regression

• Tutorial: Adding confidence bands to effect displays
• Tutorial: Plotting regression models
• Covariate adjustment in logistic mixed models: Is it worth the effort?
• Covariate adjustment in logistic regression — and some counterintuitive findings

correlational studies

• Which predictor is most important? Predictive utility vs. construct importance
• What data patterns can lie behind a correlation coefficient?
• Controlling for confounding variables in correlational research: Four caveats

mixed-effects models

• Tutorial: Adding confidence bands to effect displays
• Tutorial: Plotting regression models
• Covariate adjustment in logistic mixed models: Is it worth the effort?

cluster-randomised experiments

• Experiments with intact groups: spurious significance with improperly weighted t-tests
• Analysing experiments with intact groups: the problem and an easy solution

generalised additive models

• Interactions between continuous variables
• The problem with cutting up continuous variables and what to do when things aren't linear

non-linearities

• Interactions between continuous variables
• What data patterns can lie behind a correlation coefficient?
• Drawing a scatterplot with a non-linear trend line
• The problem with cutting up continuous variables and what to do when things aren't linear

missing data

• Causes and consequences of unequal sample sizes

open science

• The Centre for Open Science's Preregistration Challenge: Why it's relevant and some recommended background reading
• Some advantages of sharing your data and code

correction

• Experiments with intact groups: spurious significance with improperly weighted t-tests

multiple regression

• Tutorial: Adding confidence bands to effect displays
• Tutorial: Plotting regression models
• Why reported R² values are often too high

reading

• Surviving the ANOVA onslaught

tutorial

• Tutorial: Adding confidence bands to effect displays
• Tutorial: Plotting regression models
• Tutorial: Drawing a dot plot
• Tutorial: Drawing a boxplot
• Tutorial: Drawing a line chart
• Tutorial: Drawing a scatterplot

machine learning

• Classifying second-language learners as native- or non-nativelike: Don't neglect classification error rates

random forests

• Classifying second-language learners as native- or non-nativelike: Don't neglect classification error rates

R tip

• R tip: Ordering factor levels more easily

bootstrapping

• Some illustrations of bootstrapping

replication

• Draft: Replication success as predictive utility

predictive modelling

• Draft: Replication success as predictive utility